2755
Organometallics 1989,8, 2755-2759
1,5,9-Tristannacyclododecanes as Lewis Acids. Novel Structure of a Chloride Complex Klaus Jurkschat,+Henry G. Kuivila,**t Shuncheng Liu,t and Jon A. Zubietat Departments of Chemistry, Martin-Luther University Halie- Wittenberg, Postfach, DDR-40 10 Halle/S., Germn Democratic Republic, and State University of New York at Albany, Albany, New York 12222 Received March 16, 1989
A simple procedure for the preparation of substituted 1,5,9-tristannacyclododecanesRRSn(CH2)3SnRR'(CH2)3SnRR(CH2)3 (R = R' = CH3(3);R = CH,,R' = C1 (4); R = R' = Cl(5))is described. I Their complexing behavior toward chloride was investigated by means of l19Sn NMR spectroscopy. Both 4 and 5 yielded evidence for complex formation but behaved differently from each other. A complex ((5C1)(Ph3P=N=PPh3)+) (6)was isolated; its structure as determined by single-crystal X-ray diffraction shows that 5 acts as a bidentate Lewis acid via unsymmetrical transannular bridging by a chloride ion. 6 c stallizes in the monoclinic space group P2,/c with the unit-cell dimensions a = 11.691 (2) A, b = 24.652 (4)%, c = 17.353 (4) A, = 92.70 (l)', V = 5110.6 A3, and D(calcd) = 1.65 g/cm" for 2 = 4. The structure was refined to a final R value of 0.039 for 4003 independent reflections.
L
Introduction The complexation chemistry of multidentate acyclic ligands with cations has developed steadily over the last century; during the past 2 decades it has expanded rapidly to macrocyclic ligands that are a familiar part of the chemists inventory. In contrast, reports on studies of the chemistry of the complementary multidentate acids have only recently begun to appear. These include cationic species such as polyammonium compounds that interact with donors via hydrogen Apparently the earliest report that is relevant to this study involving a neutral Lewis acids is that of Shriver on ethane-1,2-bis(difluoroborane) describing evidence for a chelate e f f e ~ t . More ~ recently Katz has observed powerful bidentate effects with 1,8-naphthalenebis(dichloroborane) and a sil y l b ~ r a n e . ~Bidentate ~ complexes of 1,2-phenylenedimercury with halide ions,6aibdimethylformamide,& and tetrahydrofuransd have been isolated and characterized. Halogen-bridged and oxygen-bridged complexes of bis(halostanny1)methanes have also been isolated and characteri~ed.'*~NMR evidence has been adduced for cooperative binding of halide ions with macrocyclic and bicyclic d i t i n ~ . Transfer ~ of halide ions in liquid membrane experiments with 1,5,9-trisiladodecanehas been reported, although other evidence for complexation was not obtained.1° In this paper we report the first preparative scale synthesis of 1,5,9-tristannacyclododecanederivatives along with complexation studies with chloride of two chloro derivatives and the isolation and structure determination of the first chloride complex of a monocyclic multitin.
Scheme I
1
4
gests that these compounds form polymeric networks in the solid state by intermolecular chlorine bridges. The 'H NMR spectrum of 4 shows only a one signal for methyl protons, suggesting that it may be the all-cis isomer. The cis-cis-trans isomer would show two signals with an in(1) (a) Vogtle, F.; Weber, E. Host-Guest Complex Chemisty/Macrocycles; Springer-Verlag: Berlin, 1985. (b) Weber, E.; Vogtle, F. Nachr. Chem., Tech. Lab. 1987,35,149. (2) Pierre, J. L.; Baret, P. Bull. SOC.Chim. Fr. 1983, 11-12, II, 367. (3) Schmidtchen, F. P. Nachr. Chem., Tech. Lab. 1988, 1, 8. (4) Shriver, D. F.; Biallas, M. J. J. Am. Chem. SOC.1967, 89, 1078. (5) (a) Katz, H. E. J. Org. Chem. 1985, 50. (b) Katz, H. E. J. Am. Chem. Soc. 1985,107,1420. (c) Katz, H. E. J.Am. Chem. SOC.1986,108, 7640. (6) (a) Wuest, J. D.; Zacharie, B. Organometallics 1985, 3, 410. (b) Beauchamp, A. L.; Olivier, M. J.; Wuest, J. D.; Zacharie, B. J.Am. Chem. SOC. 1986, 108, 73. (c) Beauchamp, A. L.; Olivier, M. J.; Wueet, J. D.; Zacharie, B. Organometallics 1987,6,153. (d) Wuest, J. D.; Zacharie, B. J. Am. Chem. SOC.1987,109,4714. (7) (a) Hyde, J. R.; Karol, T. J.; Hutchinson, J. P.; Kuivila, H. G.; Zubieta, J. A. Organometallics 1982,1,404. (b) Karol, T. J.; Hutchinson, J. P.; Hyde, J. R.; Kuivila, H. G.; Zubieta, J. A. Organometallics 1983, 2, 106. (8) Jurkschat, K.; Gielen, M. Bull. SOC.Chim. Belg. 1982,91,803. (b) Chim. Belg. 1984,93,153. (c) Gielen, Gielen, M.; Jurkschat, K. Bull. SOC. M.; Jurkschat, K.; Meunier-Piret,J.; van Meerssche, M. Bull. SOC.Chim. Belg. 1984, 93, 379. (9) Newcomb, M.; Madonik, A. M.; Blanda, M. T.; Judice, J. K. Organometallics 1987,6,145. (b) Newcomb, M.; Homer, J. H.; Blanda, M. T. J. Am. Chem. SOC.1988,109,7878. (c) Newcomb, M.; Blanda, M. T. Tetrahedron Lett. 1988,29,4261. (10)Jung, M. E.; Xia, M. Tetrahedron Lett. 1988,29, 297. (11) Compound 3 has been detected previously be GC-MS but not isolated Seetz, J. W. F. L.; Schat, G.;Akkerman, 0. S.; Bickelhaupt, F. J. Am. Chem. SOC.1982, 105, 3336.
Results and Discussion As shown in Scheme I the reaction of disodium 1,3bis(trimethylstanny1)propane (1) with bis(3-chloropropy1)dimethyltin (2) yields 1,1,5,5,9,9-hexamethyl1,5,9-tristannacyclododecane(3)" in about 35% yield. Compound 3 is a colorless, volatile liquid that can be easily transformed into the trichloro derivative 4 and into the heaxachloro derivative 5 upon reaction with the appropriate number of equivalents of mercuric chloride. Compounds 4 and 5 are colorless, very fine needles that display limited solubility in noncoordinating solvents. This sugMartin-Luther University Halle-Wittenberg, *State University of New York at Albany.
0276-7333,18912308-2755%01.50/0 0 1989 American Chemical Societv , I
5
-
Jurkschat et al.
2756 Organometallics, Vol. 8, No. 12, 1989
1:l complex. Addition of only 1mol of chloride converts most of 4 to 4C1-, a symmetrically bridged species. Further - -100
Me, C,l
1-
f‘s,n/\ - - 50 6 ‘I9 Sn
-0
4CI -
- 50 - 100
0
IO CI-/Sn
20
Figure 1. Plots of chloride/tin ratios vs l19Sn chemical shifts for 4 and 5 in chloroform at 298 K (a) compound 5; (b) compound 4.
tensity ratio of 2:l. However, it is possible that both isomers are present in solution and undergo interconversion by way of a chloride exchange with inversion. Qualitative Anion Complexation Studies. The complexation of 3, 4, and 5 with chloride added as [Ph3P=N=Ph3]+C1- was studied by l19Sn NMR. Only one tin signal was observed in each experiment indicating rapid chloride exchange on the NMR time scale. When 1 equiv of chloride/tin was added to a chloroform solution of 3, a change in the chemical shift from -10.4 to -10.9 ppm occurred, indicating that no significant complexation occurred. Substantial effects were observed with 4 and 5 as can be seen from Figure 1. Although 4 is essentially insoluble in chloroform, addition of a 0.5 mol of chloride/mol of tin did cause 4 to dissolve and provide a solution showing a chemical shift of 83.1 ppm. Nonetheless it was deemed to be desirable to estimate the chemical shift of 4 itself. This could be approximated in two ways. Cyclic analogues with two methylchlorotins in rings of seven, eight, nine, and ten members show chemical shifts in the range 150-163 ppm.12* Also linear extrapolation of the plot in Figure 1 leads to a value around 150 ppm. With these values in hand one can estimate that the addition of the first 0.5 mol of chloride/mol of tin in 4 moves the signal upfield by about 70 ppm to 83.1 ppm, and the second 0.5 mol brings about an additional shift of 40-43.3 ppm. The total shift of 110 ppm suggests formation of a very stable 1:l complex. Further addition of the second and third moles of chloride caused changes in the chemical shift to 28.8 and 22.1 ppm, respectively. The values of 43.3 and 22.1 ppm fall within the range for pentacoordinate tins.13 The small increment of only 21 ppm implies little change in the average coordination number of the tins upon addition of more than 1mol of chloride/mol of 4! This may be rationalized in at least three ways. In one of these the chloride is bonded to 4 symmetrically in the complex (eq 1, S = 4). This suggests the formation of a rather stable
s + c1- sc1sc1- + c1- 2 sc12i2
4C1zS
increments simply shift the equilibrium of eq 1to the right causing the modest upfield shift observed. In a second formulation the equilibria of eq 1and 2 are established. The product of eq 1 is 4C1- and that of eq 2 is 4C12-. In each of these all three tins are pentacoordinate so the chemical shifts of the two species should be similar but not identical. The third alternative for 4C1- would involve monodentate coordination with rapid exchange of the chloride ion among the three tin sites. In this case the chemical shift should move upfield from that for 4 by one-third of the difference between tetracoordinate and pentacoordinate tins. The total change upon addition of three chlorides/mol of 4 is about 130 ppm. A shift of one-third of this value, or 43 ppm, from the initial value near 150 ppm would lead to a value of 97 ppm that is well below that of 43.3 ppm observed when 1 mol of chloride/mol of tin is present. As 5 is also insoluble in chloroform its chemical shift must be estimated, but data for the cyclic tetrachlorodistannacycloalkanes are of limited use. Observed values are as follows: for 1,1,5,5-tetrachloro-1,5-distannacyclooctane, +15 ppm (in the solid state both tins are pentacoordinate due to intra- and intermolecular chloride bridging in a linear polymeric structure12b);for 1,1,5,5tetrachloro-l,5-distannacyclononane,+119 ppm; for 1,1,6,6-tetrachloro-l,6-distannacyclodecane, +36 ppm.’& Extrapolation of the data presented in Figure 1leads to a value of about 120 ppm for 5. With use of this value the first mole of chloride causes an upfield shift of about 90 ppm to -32 ppm in the chemical shift. The second mole causes an additional shift of 91 ppm to -123 ppm, whence further additions cause a change of only about 10 ppm. The striking observation here is the large chemical shift changes caused by the additions of each of the first 2 mol of chloride; this represents a distinct difference in the the behavior of 5 as compared to that of 4. It leads to the conclusion that 5 forms very stable complexes with one and with two chlorides, respectively; Le., the equilibria of eq 1 and 2 (S = 5 ) lie very far to the right. The average coordination number is lower in 5Cl- than in 5Cl2- whose maximum value must be near five because a coordination number of six would require a chemical shift higher by at least 100 ppm.13 The structure shown for 5C1f is an em-
t 1) (2)
(12)(a) Farah, D. Ph.D. Dissertation, State University of New York a t Albany, 1987. (b) Jurkschat, K.;Liu, S., unpublished observations. (13)(a) Smith, P. J.; Tupciauskas, A. P. Annu. Rep. NMR Spectrosc. 1978,8, 291. (b) Petrosyan, V. S. B o g . Nucl. Magn. Reson. Spectrosc. 1977,11,115. (c) Wrackmeyer, B. Ann. Rep. NMR Spectrosc. 1985,16, 73.
5CI -
inently reasonable one for it accommodates the stoichiometry with two chlorides/moland the pentacoordinate structure of all of the three tins. Three structures can be considered for 5C1- in relation to chemical shifts on the assumption that the total chemical shift change from 5 to 5C1- is 250 ppm. If 5C1- is a stable, rapidly exchanging
1,5,9-Tristannacyclododecanes as Lewis Acids Table I. Atomic Coordinates (XlO') and Equivalent Isotropic Displacement Parameters (A* X loa) for the Anionic Moiety of 7 X z Y 2783 (1) -1223 (1) 10793 (1) 3957 (1) 877 (1) 11701 (1) 1634 (1) -50 (1) 9299 (1) 4447 (2) -1763 (1) 10719 (2) 2007 (3) -1758 (1) 11810 (2) 2234 (3) 12342 (2) 910 (2) 5055 (3) 1522 (1) 12393 (2) 2331 (2) 267 (1) 8135 (1) -312 (2) 406 (1) 9058 (1) 3644 (2) -566 (I) 9584 (1) 3253 (8) -554 (4) 11502 (5) 4477 (8) -360 (4) 11481 (6) 4729 (8) 123 (4) 12015 (5) 3597 (8) 1231 (4) 10634 (5) 3299 (7) 820 (4) 9991 (5) 2150 (7) 560 (3) 10103 (5) 862 (8) -811 (4) 9034 (5) -1330 (4) 1487 (9) 9155 (5) 1585 (8) -1550 (4) 9968 (5)
Sn(l)-C1(1) Sn(1)-C1(7) Sn(l)-C(9) Sn(2)-C1(4) Sn(2)-C (4) Sn(3)-C1(6) Sn(3)-C(6) C(l)-C(2) C(4)-C(5) C(7)-C(8)
Table 11. Bond Lengths (A) 2.365 (3) Sn(l)-C1(2) 2.906 (2) Sn(l)-C(l) 2.135 (9) Sn(2)-C1(3) 2.353 (3) Sn(2)-C(3) 2.111 (9) Sn(3)-C1(5) 2.556 (2) Sn(3)-C1(7) 2.140 (8) Sn(3)-C(7) 1.510 (12) C(2)-C(3) 1.554 (12) C(5)-C(6) 1.483 (12) C(8)-C(9)
2.445 (2) 2.132 (9) 2.361 (3) 2.130 (9) 2.387 (2) 2.700 (2) 2.127 (8) 1.542 (12) 1.509 (11) 1.540 (12)
monodentate structure, then the chemical shift change would be upfield by 83 ppm to +37 ppm, well below that observed. If it is a bidentate structure analogous to that shown for 4C1- the chemical shift change would be 166 ppm to -46 ppm. And if it were a tridentate species, all three tins being pentacoordinate, the shift would be to the maximum value observed, around -130 ppm. The value for the bidentate structure is clearly the only one near that observed, making it the preferred structure. I t would be desirable to obtain evidence to test these conclusions. To this end we have sought to isolate and to determine the structures of complexes and have succeeded in one case. When an attempt was made to obtain a crystalline complex of 4C1-, an oil that turned to a glass 6 resulted, but no crystals could be obtained. Reaction of 5 with Ph3P=N=PPh3+C1- provided crystals of a compound (7) with the composition 5CV. Molecular Structure of 6. Single-crystal X-ray diffraction showed the anionic moiety of 6 to have the structure shown in Figure 2. Atomic coordinates and equivalent isotropic displacement parameters for all atoms of the anion are presented in Table I. Bond lengths and angles are presented in Tables I1 and 111, respectively. The structure shows that 5 acts as a bidentate Lewis acid toward the chloride ion in 6: Sn(1) and Sn(3) are pentacoordinate whereas Sn(2) is tetracoordinate. The chloride bridge is unsymmetrical, and the trigonal-bipyramidal geometries around Sn(1) and Sn(3) are distorted, but to different degrees. At Sn(3) the equatorial groups fall in the same plane with a maximum deviation of ca 0.1 A from the best least-squares plane through the atoms although the angles are different from 120O. For example, the intraannular angle C(7)-Sn(3)-C(6) is expanded to 150.5 (3)", and the other two angles are smaller: 104.9 (3)" and 104.5 ( 2 ) O . The axial tin-chlorine bond lengths at Sn(3) are 2.700 (2) and 2.556 (1)A. At Sn(1) these lengths are
Organometallics, Vol. 8, No. 12, 1989 2757
Figure 2. ORTEP view of the structure of 7. The cationic part is omitted for clarity. Table 111. Bond Angles (deg) c1(2)-Sn(l)-Cl(l) 94.2 (1) Cl(7)-Sn(l)-Cl(l) C1(7)-Sn(l)-C1(2) 178.3 (1) C(1)-Sn(1)-Cl(1) C(l)-Sn(U-C1(2) 94.3 (3) C(l)-Sn(l)-C1(7) C(9)-Sn(l)-Cl(l) 105.4 (3) C(9)-Sn(l)-C1(2) C(g)-Sn(l)-C1(7) 86.7 (2) C(9)-Sn(1)-C(1) C1(4)-Sn(2)-C1(3) 100.5 (1) C(3)-Sn(2)-C1(3) C(3)-Sn(2)41(4) 103.9 (3) C(4)-Sn(2)-C1(3) C(4)-Sn(2)-C1(4) 105.6 (3) C(4)-Sn(2)-C(3) C1(6)-Sn(3)-C1(5) 92.8 (1) C1(7)-Sn(3)-C1(5) C1(7)-Sn(3)-C1(6) 177.5 (1) C(6)-Sn(3)-C1(5) C(6)-Sn(3)-C1(6) 91.3 (2) C(6)-Sn(3)-C1(7) C(7)-Sn(3)-C1(5) 104.9 (3) C(7)-Sn(3)-C1(6) C(7)-Sn(3)-C1(7) 88.9 (3) C(7)-Sn(3)-C(6) Sn(3)-C1(7)-Sn(l) 94.2 (1) C(2)-C(l)-Sn(l) C(3)-C(2)-C(1) 112.5 (7) C(2)-C(3)-Sn(2) C(5)-C(4)-Sn(2) 114.7 (6) C(6)-C(5)-C(4) C(5)-C(6)-Sn(3) 115.8 (6) C(8)-C(7)-Sn(3) C(9)-C(8)-C(7) 117.0 (8) C(8)-C(9)-Sn(l)
87.6 (1) 106.1 (3) 85.2 (3) 92.8 (2) 147.1 (4) 105.3 (3) 106.3 (3) 131.3 (4) 89.3 (1) 104.5 (2) 89.5 (2) 89.2 (3) 150.5 (3) 116.7 (6) 116.5 (6) 110.5 (7) 121.9 (6) 121.8 (6)
more divergent at 2.906 (2) and 2.445 (2) A; the trigonal-bipyramidal geometry is more distorted at this site. The largest of these bond lengths are comparable to those observed in the 1,2-bis(stannylethane), [ (Ph,C1SnCH2)2C1]-[Ph3P=N=PPh3]+.14 These values may also be com ared with the shorter equatorial Sn-C1 lengths of 2.387 at Sn(3) and 2.445 A at Sn(l), which are similar to those in other pentacoordinate diorganotin dic h l o r i d e ~and , ~ ~the more nearly normal values of around 2.36 A at Sn(2). The intraannular C(S)-Sn(l)-C(l) bond angle is opened up to 147.1 (4)'. Additional angle strain in the ring is found in the value of 117' for the C(7)-C(8)-C(9) bond and values near 1 2 2 O for the C(8)-C(7)Sn(3) and C(8)-C(9)-Sn(l) bonds. Further strain is observed in the C(4)-Sn(2)-C(3) bond angle of 131.3 (4)'. Thus formation of the Sn-Cl-Sn bridge occurs at the cost of significant strain in the ring angles in the six-membered ring; this implies substantial driving force for formation of the bidentate structural unit provided that 5 itself is relatively unstrained, as inspection of Dreiding models suggests. I t is rather striking that that all three tins are not involved in coordination in 5C1- as appears to be the case in 4C1-. Clarification of this point and others must await further study.
1
Experimental Section General Data. All solvents were dried according to standard procedures and freshly distilled before use. 'H,lSC, and "?9n NMR spectra were recorded on a Varian XL 300 spectrometer (14)Jurkschat, K.; Hesselbarth, S.; Tzschach, A.; Piret-Meunier,J., in press.
2758 Organometallics, Vol. 8, No. 12, 1989 Table IV. Summary of Crystal Data and Experimental Details for the Structural Study of 7 (a) Crystal Parameters at 23 O c a a, A 11.691 (2) V, A3 5110.3 (12) b, k, 24.652 (4) space group R1/c c, A 17.353 (4) Z 4 8, deg 92.70 (1) Ddd, g cm-3 1.65 (b) Measurement of Intensity Data cryst dimen, mm 0.24 X 26 X 0.21 instrument Nicolet R3m/V X = 0.71073 k, radiatn Mo K a scan mode coupled O(crystal)-e(counter) scan rate 3-20" min-' scan length [28(K~t1 - l.O]-[28(K~tz+ l.O)]" background stationary crystal, stationary counter, measurement at the beginning and end of each 28 scan, each for the time taken for the scan stds 3 collected every 197 no. of reflctns collected 6545 (*:h,+k,+l) no. of reflctns used 4003 [F,> 6(F0)] no. of parameters 391 (c) Reduction of Intensity Data and Summary of Structure Solution and Refinementb' absorptn correctn based on scans for 5 reflections with x near 90" or 270" TIMI/ Tmin 1.090 structure soln Patterson map yielded the Sn positions; all remaining non-hydrogen atoms were located via standard Fourier techniques anomalous scattering factorsC neutral atomic scattering factors were used throughout anomalous dispersnd applied for all non-hydrogen atoms final discrepancy factors' R = 0.039 R, = 0.041 goodness of fit' 1.096
+
"From a least-squares fitting of the setting angles of 25 reflections. All calculations were performed on a Data General Nova 3 computer with 32K of 16-bit words using local versions of the programs as described in: Sheldrick, G. M. Nicolet SHELXTL Operations Manual: Nicolet XRD Corp.: Cupertino, CA, 1979. 'Cromer, D. T.; Mann, J. B. Acta Crystallogr. A 1968, 24, 321. International Tables for X - r a y Crystallography: Kynoch Press: Birmingham, U.K., 1974; Vol. 111. e R = z[[iFoI- l F J / ~ l F o land ] R, = [MlF0I- l~c1)2/LlFo121'~2, where w = l/oZ(Fo)+ and g = 0.005. 'GOF = [xc.(IFol- lFc1)2/(N0- NV)]'/2, where NO is the number of observations and NV is the number of variables. 8 Data corrected for background, attenuation, Lorentz, and polarization effects. operating a t 111.86 MHz with tetramethylsilane and tetramethyltin as external references. Typical operating parameters were as follows: 50-KHz spectral width, 5-5 pulse width, 0.3-5 acquisition time, and 30-K data points. X-ray Crystallography. The details of the crystal data collection methods and refinement procedures are given in Table IV. (Supplementary Tables I-VI provide atomic positional and thermal parameters, bond lengths and angles, anisotropicthermal parameters, calculated hydrogen atom positions, and calculated and observed stmcture factors.) Full details of the crystallographic methodologies have been described.16 Extinction correction was not deemed necessary. Idealized hydrogen atom positions were used throughout the analysis, with the C-H distance set at 0.96 A. All non-hydrogen atoms of the anion and the N and P atoms of the cation were refined anisotropically. Isotropic temperature factors were employed for the ring carbons of the cation. 1,3-Bis(trimethylstannyl)propane. This compound was prepared from (trimethylstanny1)sodium and 1,3-dichloropropane (15) Jurkechat, K.; Schilling, J.; Mugge, C.; Tzschach, A.; MeunierPiret, J.; van Meersche, M.; Gielen, M.; Willem, R. Organometallics 1988,
7 , 38. (16) Nicholson, T.; Zubieta, J. A. Polyhedron 1988, 7, 171.
Jurkschat et al. in liquid ammonia in a procedure similar to that for the prepabp 110 "C (10 Torr); ration of 2,2-bis(trimethyl~tannyl)propane:~ 'H NMR (CDCl,) 6 0.04 (8, 18 H, 2J(11DSn-C-H) = 52.2 Hz, SnMeJ, 0.85 (t,4, 2J(110Sn-C-H) = 51.5 Hz, SnCHJ, 1.72 (9, 2 H, CHz). 1,3-Bis(chlorodimethylstannyl)propane ( 1). 1,3-Bis(trimethlylstanny1)propane (11.1 g, 0.03 mol) and dimethyldichlorostannane (13.2 g, 0.06 mol) were heated for 12 h at 60 "C. The trimethylchlorostannane formed was removed in vacuo and the residue recrystallized from diethyl ether to yield 10.5 g (85.3%) of the desired product: mp 81 "C; 'H NMR (CDClJ 6 0.64 (s, 12 H, 2J(11%n-C-H) = 55 Hz, SnCHd, 1.37 (t, 4 H, 2J(11%nC-H) = 50.6 Hz), 2.06 (q, 2 H, 3J(110Sn-C-C-H) = 68.9 Hz, CH,).
1,1,5,5,9,9-Hexamethyl-1,5,9-tristannacyclododecane. Bis(2-(methoxycarbony1)ethyl)dimethylstannane (217 g, 0.67 mol) was added dropwise to a suspension of LiAIHl (28 g, 0.738 mol) in 1L of diethyl ether. The mixture was refluxed for 1h. Under ice cooling, 200 mL of water was added and the mixture was filtered. The ether layer was dried over MgSO,, and, after the solvent was evaporated off, the product was distilled to give 115 g (43%) of bis(3-hydroxypropy1)dimethylstannaneas a colorless oil: bp 120-125 "C 0.05 Torr); 'H NMR (CDC13)6 -0.01 (s,6 H, 2J(11%n-C-H) = 51.1 Hz),0.76 (t,4 H, SnCHJ, 1.70 (q,4 H, CHa, 3.52 (t, 4 H, CH2),2.25 (8, 2 H, OH); 'lgSn NMR (CDCI,) 6 3.04. To a magnetically stirred solution of the bis(3-hydroxypropyl)dimethylstannane (47 g, 0.176 mol) in 200 mL of methylene chloride and 55 g of tetrachloromethane was added triphenylphosphine (92.2 g, 0.352 mol) in small portions. The mixture was stirred for 24 h, and the solvent was evaporated in vacuo to yield an oil. Diethyl ether was added, and the resulting mixture was stored at -10 "C in order to cause precipitation of triphenylphosphine and triphenylphosphine oxide. After filtration, evaporation of the solvent and repeated fractionation 10 g (18.7%) of bis(3-chloropropy1)dimethylstannane (2) was obtained bp 88 "C (0.01 Torr); 'H NMR (CDC13)6 0.07 (s, 6 H, 2J(11SSn-C-H) = 52.3 Hz, Sn(CH3)z),0.90 (t, 4 H, SnCH,), 1.91 (4,4 H, CH,), 3.47 (t, 4 H, CH,Cl); 'lOSn NMR (CDC13) 6 5.5. To a solution of 1,3-bis(chlorodimethylstannyl)propane (12.3 g, 0.03 mol) in 1 L of tetrahydrofuran (THF) and 1 L of liquid ammonia was added 2.76 g (0.12 mol) of sodium in small quantities. The mixture was stirred for 30 min whence it turned yellow. To this solution was added 2 (9.2 g, 0.03 mol) in 40 mL of THF dropwise at -78 "C over a period of 1h. Then 300 mL of diethyl ether and 100 mL of water were added, and the organic layer was separated and dried over MgSO,. After evaporation of the solvent the residue was fractionated in vacuo to yield 6 g (35%) of 1,1,5,5,9,9-hexamethyl-l,5,9-tristannadododecane (3): bp 134-136 "C (0.02 Torr); 'H NMR (CDC13) 6 0.02 (s, 18, 2J(11gSn-C-H)= 50.0 Hz, Sn(CH3),), 0.90 (t, 12 H, 2J(11gSn-C-H) = 49.8 Hz, SnCHJ, 1.80 (q,6 H, 3J(11%n-C-C-H) = 54.7 Hz,CHJ; '% NMR (CDC13) 6 -10.6 (1J(11gSn-13C)= 299.7 Hz, SnCH,), 16.14 ('J(119Sn-13C)= 339.9 Hz, 3J(110Sn-C-C-13C) = 53.2 Hz, SnCH2), 24.85 (2J(110Sn-C-13C)= 18.3 Hz, CH,); 'l9Sn NMR (CDCl,) 6 -10.4. Anal. Calcd for Cl6HSSn3: C, 31.46; H, 6.29. Found C, 31.55; H, 6.54. 1,5,9-Trichloro-1,5,9-trimethyl1,5,9-tristannacyclododecane (4). To a solution of 2.5 g (4.36 mmol) of 3 in 30 mL of acetone was added mercuric chloride (3.55 g, 13.080 mmol) under magnetic stirring at 0 "C. The mixture was stirred for 30 min at this temperature and for 1h at room temperature. Then the solvent was evaporated,and the byproduct CHJ-IgCl was removed at 0.02 Torr and 90 "C. The residue was transferred into a Soxhlet extractor and extracted with methylene chloride. The extract provided 2.1 g (76%) of 4: mp 224-226 "C; 'H NMR (CDCl,) 6 0.63 (s, 9 H, 2J(110Sn-C-H) = 52.7 Hz, SnCH3), 1.45 (t, 12 H, SnCH,), 2.13 (q,6 H, 3J(11gSn-C-C-H) = 68.6 Hz,CH,). Anal. Calcd for Cl2HZ,Cl3Sn3:C, 22.73; H, 4.26. Found: C, 22.62; H, 4.23. 1,1,5,5,9,9-Hexachloro-l,5,Stristannacyclododecane(5). To of 3 in 50 mL of acetone was added a solution of 1.87 g (3.28 "01) mercuric chloride (5.34 g, 19.7 mmol) under magnetic stirring at 0 "C. The mixture was stirred for 15 min at this temperature and then refluxed for 48 h. The solvent was evaporated, and the byproduct CH3HgCl was removed at 0.02 Torr and 90 "C. The residue was placed in a Soxhlet extractor and extracted with methylene chloride. The extract provided 1.5 g (66%) of 5: mp
Organometallics 1989,8, 2759-2766 166-168 OC; 'H NMR (CDClJ 6 1.97 (t, 12 H, SnCH,), 2.37 (q, 6 H, CHJ. Anal. Calcd for CJ-I18Cl&h3: C, 15.5;H, 2.59. Found C, 15.32;H, 2.72. Complex of 4 with [Ph+N=PPh3]+Cl- (6). Into a solution of 4 (0.127g, 0.2 mol) and [Ph*N=PPh,]+Cl(0.115g, 0.2 mol) in 3 mL of methylene chloride was added 3 mL of hexane. Slow evaporation led to precipitation of an oily product that solidified to a glasslike product: 'H NMR (CDC13)6 0.70 (s, 9 H, ,J('lSSn-C-H) = 59.6 Hz, SnCH,), 1.65 (t, 12 H, 2J(11gSn-C-H)= 62.0Hz, SnCHJ, 2.38 (q,6H, 3J(11%n-C-C-H) = 110.7 Hz,CH,), 7.70 (m, 12 H, o-Ph), 7.53 (m, 18 H, m- and p-Ph); 13C NMR (CHC13) 6 1.68 (1J(11gSn-13C)= 389.3 Hz, SnCH,), 28.36 ('J(11sSn-13C)= 456.6 Hz, 3J(11gSn-C-C-13C) = 49.6 Hz, SnCH,), 22.38 (2J(11BSn-C-'3C)= 27.0 Hz, CH2). Complex of 5 with [Ph3P=N=PPh3]+Cl- (7). To 3 mL of methylene chloride and 2 mL of hexane were added 78.0 mg (0.136 "01) of [Ph+N=Ph3]+Cl- and 95.4mg (0.136mL) of 5. Slow evaporation of the solvent yielded 155 mg (90%) of colorless crystals: mp 224-225 OC; 'H NMR 6 2.16 (t, 12 H, 2J(11gSn-C-H) = 71.3 Hz,SnCH2),2.59 (q, H, 3J(11%n-C-C-H)= 187.7 Hz,CHJ, 7.46 (m,18 H, m- and p-Ph), 7.65 (m, 12 H, o-Ph); 13C NMR (CDC13)6 38.65 (lJ(llgSn-lSC)= 592.6 Hz, 3J(11sSn-C-C-C) =
2759
50.1 Hz SnCH,), 21.91 (2J(11gSn-C-C) = 41.6 Hz, CH,). Anal. Calcd for C45H48NP2Sn3; C, 42.58;H, 3.78. Found C, 42.42;H, 3.71.
Acknowledgment. K.J. thanks the International Research and Exchanges Board (Princeton, NJ) for a grant. We also thank Professor Shelton Bank for valuable suggestions and the reviewers for helpful criticisms. Registry No. 1, 123126-09-0; 2, 123126-05-6; 3,85443-05-6; 4, 123126-06-7; 5, 123126-07-8; 6, 123126-11-4;7, 123126-13-6; [Ph3P=N=PPh3]+Cl-, 21050-13-5;1,3-bis(trimethylstnyl)propane, 35434-81-2;(trimethylstannyl)sodium,16643-09-7;1,3dichloropropane, 142-28-9;dimethyldichlorostannane, 753-73-1; bis((2-methoxycarbonyl)ethyl)dimethylstannane,115152-95-9; bis(3-hydroxypropyl)dimethylstannane,123126-08-9.
Supplementary Material Available: Tables of atomic coordinates and equivalent isotropic displacement parameters, bond distances, bond angles, and H-atom coordinates and isotropic displacement parameters (6 pages); a listing of observed and calculated structure factors (15pages). Ordering information is given on any current masthead page.
Electronic Absorption Spectra of Diorganogermylenes in Matrices: Formation of Diorganogermylene Complexes with Heteroatom-ContainingSubstrates Wataru Ando, Hiroyuki Itoh, and Takeshi Tsumuraya Department of Chemistty, The University of Tsukuba, Tsukuba, Ibaraki 305, Japan Received March 31. 1989
Diorganogermylenes were generated in hydrocarbon matrices at 77 K by the photolysis of 7germanorbornadienes la-e or bis(trimethylsily1)germanes 2a-g. The germylenes show electronicabsorption bands at 420-558 nm. The germylenes react with heteroatom-containing substrates (bo,RzS,R3P,R3N, RCl, and ROH) to form adducts, which show characteristic absorption bands at shorter wavelengths than those of germylenes.
Introduction Divalent compounds of elements of group 14 have been the subject of considerable interest in recent y e a r ~ . l - ~ Although diorganogermylenes have been postulated as reactive intermediates in many reactions,H spectroscopic
Scheme I 8
P'
1 a:R=R' =Me
(1) Jones, M., Jr., Moss, R. A., Eds. Reactioe Intermediates; WileyInterscience: New York 1978, Vol. 1; 1981, Vol. 2; 1985, Vol. 3. (2) SatgB, J.; Massol, M.; Riviere, P. J. Organomet. Chem. 1973,56, 1. (3) Neumann, W. P. In The Organometallic and Coordinution Chemistry of Germanium, Tin, and Lead; Gielen, M., Harrison, P. G., Eds.; Freund: Tel Aviv, 1978. (4) Mars, R.; Neumann, W. P.; Hillner, K. Tetrahedron Lett. 1984,25, 625. (5) Ma, E. C.-L.; Kobayashi, K.; Barzilai, M. W.; Gaspar, P. P. J. Organomet. Chem. 1982,224, C13. (6) (a) Schriewer, M.; Neumann, W. P. J. Am. Chem. SOC.1983,105, 897. (b) K h h e r , J.; Neumann, W. P. J. Am. Chem. SOC.1984,106,3861. (c) K h h e r , J.: Neumann. W. P. Ormnometallics 1985.4. 400. (d) Michela, E.; Neumann, W. P. Tetrahe-dron Lett. 1986,27, 2455. ( e ) Neumann, W. P.; Michels, E.; K h h e r , J. Tetrahedron Lett. 1987,28, 3783. (f)Billeb, G.; Neumann, W. P.; Steinhoff, G. Tetrahedron Lett. 1988,29, 5245. (7) Barrau, J.; El Amine, M.; Rima, G.; Satg6, J. J. Organomet. Chem. 1984,277,323.
(8) (a) Ando, W.; Tsumuraya, T.; Sekiguchi, A. Tetrahedron Lett.
1985,26,4523. (b) Ando, W.; Tsumuraya, T. Tetrahedron Lett. 1986,27, 3251. (c) Ando, W.; Tsumuraya, T.; Goto, M. Tetrahedron Lett. 1986, 27,5105. (d) Tsumuraya, T.; Sato, S.; Ando, W. organometallics 1989, 8, 161. (e) Ando, W.; Tsumuraya, T. Organometallics 1989,8,1467. (0 Ando, W.; Tsumuraya, T. J. Chem. SOC.,Chem. Commun. 1989, 770.
0276-7333/89/2308-2759$01.50 f0
b:R=R'=Et C:R=R~="B~
d:R=?h,
R'=Me
e:R=R'=Ph
2a:R=R'=?h b: R-R ' = Ta 1 (4-me thy1 phenyl ) c:R=Mes(2,4,6-trimethylphenyl), d:R=R'=Xy(2,6-dimethylphenyl) e:R=R'=Ar(2,6-diethylphenyl)
R'=t8u
f:R=R'=Mes
g:R=R1=Ar'(2,4,6-tr~isopropylphenyl)
data on diorganogermylenes remain rather limited. Dimethylgermylene and some other dialkyl- and diarylgermylenes have been observed by UV s p e c t r ~ s c o p y ~ ~ J ~ ~ ~ (9) (a) Kira, M.; Sakamoto, K.; Sakurai, H. J. Am. Chem. SOC.1983, 105, 7469. (b) Sakurai, H.; Sakamoto, K.; Kira, M. Chem. Lett. 1984, 1379.
1989 American Chemical Society
